1 What is rabbit? 2 rabbit.engr.oregonstate.edu NVIDIA Titan Black rabbit.engr.oregonstate.edu PCIe Bus PCIe Bus 15 SMs 2 E5-2630 Xeon Processors 2880 CUDA cores 16 Cores total 6 GB of memory 64 GB of memory 2 TB of disk OpenCL support CUDA support Mike Bailey [email protected] This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License Computer Graphics rabbit.pptx mjb – April 16, 2019 Computer Graphics mjb – April 16, 2019 What is rabbit? 3 What is rabbit? 4 rabbit 151% lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit rabbit lives in a rack in our server room in the Kelley Engineering Center: Byte Order: Little Endian CPU(s): 32 On-line CPU(s) list: 0-31 Thread(s) per core: 2 Core(s) per socket: 8 Socket(s): 2 NUMA node(s): 2 Vendor ID: GenuineIntel rabbit CPU family: 6 Model: 63 2 TB of disk Stepping: 2 CPU MHz: 2399.982 BogoMIPS: 4799.30 Virtualization: VT-x L1d cache: 32K L1i cache: 32K L2 cache: 256K L3 cache: 20480K NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30 NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31 Computer Graphics mjb – April 16, 2019 Computer Graphics mjb – April 16, 2019 What is rabbit? 5 What is rabbit? 6 CPU NVIDIA Titan Black Titan rabbit ports CPU PCIe bus and riser network rabbit mgmt port Computer Graphics mjb – April 16, 2019 Computer Graphics mjb – April 16, 2019 1 Getting to rabbit and setting up your account 7 Compiling and running C/C++ on rabbit 8 Lowercase letter ‘L’ To login to rabbit: ssh rabbit.engr.oregonstate.edu –l yourengrusername icpc -o try try.cpp -lm -openmp -align -qopt-report=3 -qopt-report-phase=vec Put this in your rabbit account’s .cshrc : or setenv INTEL_LICENSE_FILE [email protected] setenv ICCPATH /nfs/guille/a2/rh80apps/intel/studio.2013-sp1/composer_xe_2015/bin/ set path=( $path $ICCPATH ) g++ -o try try.cpp -lm -fopenmp source /nfs/guille/a2/rh80apps/intel/studio.2013-sp1/bin/iccvars.csh intel64 setenv CUDA_PATH /usr/local/apps/cuda/cuda-9.2 setenv LD_LIBRARY_PATH $CUDA_PATH/lib64:$LD_LIBRARY_PATH set path = ( $path $CUDA_PATH/bin ) Then activate these values like this: source .cshrc (These will be activated automatically the next time you login.) Computer Graphics mjb – April 16, 2019 Computer Graphics mjb – April 16, 2019 Compiling for OpenCL 9 Compiling for CUDA 10 printinfo: printinfo.cpp arrayMul: arrayMul.cu icpc -o printinfo printinfo.cpp /usr/lib64/libOpenCL.so -lm -openmp nvcc -o arrayMul arrayMul.cu Computer Graphics mjb – April 16, 2019 Computer Graphics mjb – April 16, 2019 The printinfo Program Output 11 Reservation System – Please use It!! 12 Number of Platforms = 1 Platform #0: https://secure.engr.oregonstate.edu/engr/resources/bailey Name = 'NVIDIA CUDA' Vendor = 'NVIDIA Corporation' Version = 'OpenCL 1.1 CUDA 7.0.18' Profile = 'FULL_PROFILE' Device #0: Type = 0x0004 = CL_DEVICE_TYPE_GPU 15*192 = 2880 CUDA cores! Device Vendor ID = 0x10de (NVIDIA) Device Maximum Compute Units = 15 Device Maximum Work Item Dimensions = 3 Device Maximum Work Item Sizes = 1024 x 1024 x 64 Device Maximum Work Group Size = 1024 Device Maximum Clock Frequency = 1071 MHz Device Extensions: cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 Computer Graphics mjb – April 16, 2019 Computer Graphics mjb – April 16, 2019 2.